A prediction model of Network traffic noise reduction based on PSO-VMD and BiLSTM

Network traffic is non-linear and non-stationary in nature, which is difficult to predict. In this paper, to solve this problem, a network traffic prediction method based on the combined PSO-VMD-BiLSTM model is proposed. Firstly, the parameters of the variational modal decomposition (VMD) model are optimized by the particle swarm optimization (PSO) model. Secondly, the PSO-VMD model decomposes and reconstructs the network traffic data to reduce the noise in the original traffic data. Then the bi-directional long and short-term memory (BiLSTM) model is used to predict the reconstructed high-frequency, mid-frequency, and low-frequency sequences, respectively. Finally, the predicted values are summed up. This paper demonstrates that the combined PSO-VMD-BiLSTM model has good adaptability and prediction effect through simulations and comparative experiments on actual network traffic data.